Helical polymers are formed by many proteins found in bacterial, archaeal and eukaryotic cells, and can also be present as viral capsids, nucleocapsids and tails. In certain cases most of the protein found in a cell is in the form of a helical polymer, so methods to study the structure and dynamics of such polymers have great general interest. Particular helical complexes, such as the pili and flagellar filaments of pathogenic bacteria, have a very immediate relevance to human disease. We propose to further develop, extend, apply and support new methods for the three-dimensional reconstruction of such polymers from electron microscopic images. Our work in this area has already had a large impact on a number of NIH-supported projects from many laboratories, but we are now at a transition point where the potential for achieving high- resolution structures from numerous samples is now quite high. Given the imminent arrival of a Titan Krios TEM we need support to further develop our Iterative Helical Real Space Reconstruction approach and optimize the processing of large numbers of high-resolution images of polymers. Effort will be invested in using new algorithms for alignment and reconstruction, as well as in developing methods that make use of prior knowledge about the spatial relations between different segments that have been cut from the same filament. All of the development work will use samples that have great interest to a large community and that have a direct relation to human health. These range from bacterial pili to viral capsids to the protein product of an oncogene. We have established that processing of such images can scale with the number of processors, so effort will be invested in making these programs easy to use on relatively inexpensive and commercially-available clusters. Tools for detecting potential ambiguities in helical symmetry will be developed, but the main tool will be bringing the resolution of helical reconstructions to the point where secondary structure can be recognized. At this resolution these ambiguities disappear.